Specifically, no. As breakout analyst Ethan Chen of DeCalibrate (to which you should subscribe) has said, early vote analysis tells you zero about persuasion, i.e. the changes of individual voters beyond imputed past partisan behavior. As an example, in North Carolina in 2024, Republican President Donald Trump won the state by 3 points, while Democratic Gov. Josh Stein won by 15. The same early voters contributed to this 18-point ticket split. No pre-election look at who is voting could have estimated that delta. Nevertheless, I still wanted to build on my 2024 early vote analysis of Cameron County (where I lived at the time).

Using voter file data for 24 counties representing a cross-section of Texas’s demographics, I geocoded every voter in the lists and cross-referenced census geography to model race for individual voters. Using Catalist’s voterfile-validated 2024 post-election data, I used the gathered demographic data to create partisan priors that were blended with precinct geographic data and primary participation data in order to model 2024 voter partisanship.

Using 2024 partisanship for the voters, I aimed to model the “recalled 2024 preference” of the voters showing up who cast ballots two years ago. This question, being validated by Catalist data, is infinitely easier to measure than attempting to guess 2026 persuasion. Measuring recall would give the yardstick required for Democrats to hit 50%. A closer recalled electorate would imply fewer GOP voters required to switch over to make a Democratic victory possible.

After tracking 12 days of early voting data that got steadily more Republican, the pool of 2024 voters would have voted for Trump by around 9.7. This was roughly 4 points of turnout differential from 2024, when the state voted for Trump by 13.7. Pre-election polling that showed Talarico ahead in Trump +14 electorates implied that a Trump +9.7 electorate should favor him as well.

When estimating the Election Day surge, Texas’s Tuesday vote usually accounts for 20 to 35 percent of the total ballots cast. The number is usually larger in midterms. The Republican sources I talked to for guidance on crafting assumptions helped steelman a conservative case, and estimated 2.7 million Election Day voters. These voters would, in theory, be far more Republican than the younger, minority-heavy electorate that showed up during early voting. Based on 2022 and 2024, Election Day voting shifted the early vote result a few points towards the GOP. If this pattern manifested in 2026, given the relatively lackluster GOP turnout, I figured it would be a similar, but diminished shift. I thus guessed the final post-election electorate would have voted for Trump by 11 points among 2024 voters.

Based on this, I had to estimate how much persuasion Democrats would really obtain out of this electorate. Given the observed phenomenon of younger and nonwhite Trump voters who respond to polls lying about their vote choice, polls were underestimating the shift away from the GOP. Out of an abundance of caution, I dampened these persuasion effects, to protect against non-response bias from shy hard-right Republican voters that do not answer polls. Thus, the 14-point persuasion effects seen in the polls went down to something closer to 7.

Factoring in the 1.2 million or so voters who did not show up in 2024, the picture changed. Given that these people skewed overwhelmingly young, nonwhite, and female, and had overwhelmingly Democratic primary ballots in 2026, they likely would add 3-4 points of margin to the Democratic ticket. I thus said that Talarico would win by 1, and Abbott would win by 3.

Of course, Talarico won by 12, and Hinojosa won by 7. So my own guess about persuasion was wrong, as was everyone else’s. No serious analyst would have guessed that Democrats would win a Texas Senate election by double digits, and carry every statewide Democrat to a sweep.

No serious analyst would have predicted that low-propensity minorities swing massively to Democrats after they bolted to Trump in the previous two presidential elections. Even if you assume non-2024 voters broke proportionally to their primary participation (78% Democratic, 22% Republican), the 2024 electorate would have to swing 17 points on persuasion to match Talarico’s actual margin. The past 10 years’ hyper-polarized electorates made such swings unthinkable on this scale.

So what then, did the early vote analysis get right? Quite a bit, in fact.

First, the temporal early vote changing was something the analysis caught onto very well. Using 2024 as a benchmark, I expected the electorate to blueshift through the 12 days of the early voting period. However, unlike two years ago when Republicans had the enthusiasm edge, Democrats front-loaded their most dependable voters. The GOP picked up strength in the remaining days.

This carried further into Election Day. In 2024, suburban counties like Montgomery, Denton, and Tarrant had a blueshift from early voting to Election Day. Latino counties, on the other hand, shifted redder. In 2026, Election Day was redder in 205 out of 254 counties. During both years, latter days of the early voting period saw more lower-propensity minorities come out. As these people disproportionately swung to Trump, the 2024 recall estimate was redder than their implied 2026 partisanship was likely to be.

The early vote correctly diagnosed that rural GOP turnout was flagging, and that lower-propensity conservative voters were not showing up. Republican voters who voted in the past 3 general elections (2020, 2022, and 2024) were matching Democrats who did the same. For less frequent voters, the early voter sample was 6-7 points more pro-Harris than the equivalent in 2024.

The GOP elected officials in Texas, including Sen. Ted Cruz, all saw similar figures in their own data. Day after day, they warned that weak low-engagement GOP turnout would result in a Democratic sweep. Out of caution, I conservatively assumed the low-engagement MAGA wave would flock to the polls on Election Day.

Only 2.1 million people cast ballots on Election Day, 21% of the total. Unlike past early voting models, like the Georgia Runoff Model so expertly crafted by Lakshya Jain, there is no benchmark to compare Election Day turnout to. I had to use a guess, and I picked a deliberately conservative one that didn’t materialize: 600,000 fewer voters than I guessed showed up.

The early vote also showed that younger and nonwhite voters were punching dramatically above their weight. As an example, the Latino precincts in Harris County (Houston) turned out at higher rates relative to 2024 than working-class white areas. In Texas politics, this is unheard of. It also showed the dramatic growth in voters who didn’t participate in 2024, predominantly in Democratic-leaning and minority-heavy areas.

For the question most people are interested in, i.e., who will win, it can’t directly tell you. It can tell you a lot about who is motivated, and how likely enthusiasm gaps are to close.

As a test case of the recall question, I experimented with estimating the voter pool that showed up to a similarly competitive election in a red district: the Senate District 9 Special Election.

In northwest Tarrant County, SD-09 has been red for decades, taking in a portion of Fort Worth and its suburbs. In the special election after Sen. Kelly Hancock became State Comptroller, three candidates ran. The two Republicans were former Southlake mayor John Huffman and MAGA activist Leigh Wambsganss, and the Democrats picked Air Force veteran Taylor Rehmet as their nominee.

Given that Trump won this seat by 17, the GOP was easily expected to hold the seat. But GOP infighting between Wambsganss and Huffman mirrored the ongoing wider civil war among Republicans in the state. Rehmet got 47% in the first round, and then increased his total share in the runoff to 57%. Not only did Democrats flip the seat, they blew the doors off. The question had to be asked then, how much of this was due to turnout?

Using the early vote analysis, the 2024 voters that showed up to vote in the runoff would have supported Trump by 20.4 points - making the electorate redder than the 2024 result. This means that Rehmet won a significant portion of red voters. If you assume that the 6% of the electorate that didn’t vote in 2024 all went to Rehmet (an assumption to calculate a lower bound for crossover), then 24% of the Trump voters are implied to have voted for Rehmet. The actual estimate I will defend is closer to 28-29%.

A quarter of one party’s voters defecting is unheard of. But last Tuesday, James Talarico won SD-09 by 8 - a 25 point shift. Talarico beat Ken Paxton, who, like Wambsganss, typifies the far-right faction of the GOP that is alienating traditional conservative voters. Municipalities like White Settlement, Keller, and Watauga all voted Democratic this time, mirroring Rehmet’s special election coalition.

Having spent so much time looking at the data, Tuesday’s result should have been far easier to convey. What Taylor Rehmet achieved in Tarrant, James Talarico, Gina Hinojosa, and the Democratic slate achieved across the other 253 counties of the state as well. November 3 was the Leigh Wambsganss Election - enough erstwhile Republican voters teamed up with energized Democrats to deal far-right Republicanism a grievous blow in the state.

Concluding this early vote analysis, I believe that looking at pre-Election Day votes is useful if you have well-formed priors about how voters behave. Simply looking at turnout data by precincts will not suffice - since there are often intra-precinct turnout differentials by race and party one cannot directly observe.

Given my lack of a statewide voter file, I think it comfortably narrated the story of the 2026 election, and helped color the range of possible outcomes. In 2028 and beyond, I think a better-developed method for early vote analysis is possible with a more robust voter file. I look forward to the Texas Secretary of State and various counties providing their Election Day rosters to aid in conducting after-action post-mortems of the individual votes.

The year 1945 was the end of an age. The old world was made anew, with the lives of millions cast off as slag in the furnace. The nationalism that had grown worldwide for a century and a half culminated in the destruction of multi-ethnic communities the world over. What happened in 1945 in Europe would be repeated in 1947 in the Indian subcontinent and in 1948 in the Middle East, as the world went through its growing pains.

80 years ago, as part of the largest population transfer in history, millions of people were displaced and expelled from their homes in Central and Eastern Europe. Most visibly, a whole cultural region was wiped off the map: the land of East Prussia.

East Prussia’s rule came as all of Germany’s neighbors had numerous scores to settle. Centuries of German rule led to non-German identities carrying the penalties of discrimination and death1. First through the colonization of the Baltic Prussians, and then by the suppression of the Poles and the forcible assimilation of the remaining minorities, the steady role of Germanization culminated in 12 years of Nazi terror and genocidal warfare.

This bloody chapter of world history invited the Soviet Red Army to knock down the gates of Prussia in late 1944 and early 1945. While the Soviets and later Polish administrations of East Prussian territories were not explicitly genocidal in their intent, they too crushed the autochthonous cultures into the mortar of their new nationalist projects. Before World War II, Germans did this to Slavic and Baltic peoples - after World War II, the fortunes reversed.

Today, East Prussia’s pre-1945 inhabitants are all nearly deceased or living in modern Germany. Today, the land is home to new arrivals. The northern Russian half is now the Kaliningrad Oblast, a veritable Irkutsk-on-the-Baltic - and the southern part is the better-preserved Warmian-Masurian Voivodeship of Poland, where the old cultures have died out with time and later voluntary migrations.

80 years after the expulsions, with the story of East Prussia fast fading from living memory, this tale stands to be lost. The people, place, and communities that once made up this region would be erased. But, this project serves as an epitaph for the people and place that were East Prussia.

For the first time, we present the full geographic dataset of East Prussia, down to the individual community level, based on the German Imperial Census of 1905. East Prussia was split into three Regierungsbezirke, or administrative districts - Königsberg, Gumbinnen, and Allenstein. For the sake of standardization, all of East Prussia in this publication is based upon 1905 data, even though Allenstein district has 1910 data available.

This analysis is aimed to present an honest picture of the cultural and linguistic situation without favor to any particular nationalist inclination. As such, the Germanization efforts in the late 19th century are a concern when viewing census results. The unification of the German Empire in 1871 led to state-sponsored assimilation efforts that did not have historical precedent. These efforts drastically increased the amount of German speakers in census counts.

Many would point to the Ostsiedlung as historical precedent for this increase in Germans in the East. However, the Ostsiedlung refers to a medieval migration pattern of peoples before the rise of modern nationalism2. Given that prior to the 19th century, language and nationhood were not as strongly tied, it is very difficult to extricate these two variables from each other.

In assessing disparate groups of people that inhabit East Prussia, the base point of divergence becomes the start of the Drang nach Osten, which entered full swing in the late-19th century as right-wing German nationalists sought to co-opt medieval settlement and migration patterns in line with contemporary colonialist sentiment. At the same time, many Polish citizens in the eastern territories of Germany moved to western Germany in search of better economic opportunities3. While some of the increase in German language is undoubtedly organic and the product of Protestant Poles and Lithuanians assimilating into German identity, it is unlikely that the overall composition of people changed that much in only 30 years.

These points would both inflate the presence of Germanness in parts of East Prussia, that this historical analysis will attempt to undo. This is particularly an issue in southern East Prussia near the Allenstein District, where the situation of the Masurians has not been easy to define. A full exegesis of the Polish-Masurian-German linguistic situation will feature later in this article.

Nationalist agendas can often portray a wildly different portrait of the same scenario. German census figures show vast tracts of land inhabited by only Germans. By contrast, Polish mapmakers tend to combine groups to inflate the number of Poles. Below is a map by Jakob Spett showing the linguistic situation in eastern German territory at the time - with only two categories, German and Polish4. This categorization conforms more to Polish nationalist viewpoints of the prewar situation. All of the green areas may have spoken West Slavic languages, but East Prussian Masurians, Kashubians, and Silesians all had different motivations and cultural backgrounds. This map was one of many used in drawing the more maximalist boundaries in favor of Poland after World War I.

As it relates to East Prussia, the Spett map does not create an accurate picture. Even though the Masurian population did speak a particular dialect of Polish, the Masurian population did not strongly identify with Poland. In fact, after World War I, the plebiscite held in the Masurian region showed an overwhelming desire to remain in Germany rather than join Poland5. While some of the lopsided margin can be reduced to intimidation, terror, and fraud, the historical consensus is that a fair election would have still resoundingly ended in the pro-German side’s favor. Even later, the National Socialist Party’s strongest results in the Weimar Republic era came from the Masurian districts6. Masuria was in fact one of the only regions where the Nazis achieved an actual majority of the votes cast.

The old cultural ties that various peoples had to the German political states are now gone. Though the land is the same, the people are not at all. It is not dissimilar to the Spanish conquest of Tenochtitlan, the great Aztec capital. The Spanish cast down the city in ruins, and founded Mexico City in its place. In the Lithuanian and Polish portions of East Prussia, the place names are mostly the same, though the people are different. Most prominently though, in the Russian section, the old German capital, Königsberg, now lies in ruins, with modern Russian Kaliningrad built on top. Hundreds of villages lie abandoned throughout the region as forests cover them. Graveyards lie hidden, with weeds covering any trace of the lives that once were lived nearby.

Of course, given the high support of the East Prussian population, German and non-German alike, for the Nazi Party and other right-wing German nationalist parties before them, many would describe the expulsions as “deserved”. While this historical analysis should avoid subjective value judgments - it remains that ethnic cleansing is an act of evil, regardless of victim. The political situation at the time warranted expulsions for security reasons. At the same time, it is also an inherently unjust act.

Today, we see ethnic cleansing play out on television and the supporters of these actions use politics to justify expulsion. This analysis is an explanation and a survey of the land and people that were once East Prussia, wrecked by the devastation of World War II.

II. Geography and History

East Prussia was an isolated, poor, and rural region. Known for its dark forests and calm lakes, the region was criss-crossed by many rivers, most notably the Memel, Pregel, Inster, and Alle7. The most prominent geographical features on a map are the two spits of land, the Curonian and the Vistula Spit, which are joined at the Sambian Peninsula, famous for its numerous amber deposits.

The largest city was Königsberg, an influential pillar of German culture. While not as trafficked as other Baltic German ports such as Danzig and Stettin, the city was a key hub for inter-Baltic trade. Thinkers like Immanuel Kant and mathematicians like David Hilbert once were stalwarts of the local university, the Albertina. So great was Kant’s contribution to the world of philosophical thought that even as the entire old city of Königsberg lay in ruins, the Russian authorities have left Kant’s tomb in Königsberg still undisturbed. While overwhelmingly Protestant due to the Teutonic Order’s conversion in the wake of the Reformation, East Prussia had sizable Catholic populations in Warmia, and notable Jewish communities, particularly in the bigger cities.

Politically, the region leaned strongly conservative. State-mandated elevation of German culture over the Baltic and Slavic cultures also in the region, as well as landowning Junker prominence kept East Prussia highly reactionary. Serfdom was only abolished in 1807, and landholdings were still wildly unequal well into the 20th century, with the countryside standing dotted by sprawling castle estates.

By the turn of the century, all of East Prussia outside of heavily-Catholic Warmia voted for the German Conservative Party (DKP). Only occasional challenges from the Lithuanian Party in the Lithuania Minor region, and occasional center-left upsets prove that this region was a conservative bastion. Even post-war in the Weimar Republic era, East Prussia remained a stalwart region for the far-right DNVP, and later the Nazi Party. By the 1920s, Masurian-ness as a separate identity from Germandom almost completely died out, and the Masurian lakes saw some of the strongest outright majorities for the NSDAP anywhere in Germany.

Politically, the German province of East Prussia descends from the 1795 Partition of Poland. Then, the land of Warmia, also known as Royal Prussia, was added to Ducal Prussia during the dismemberment of Poland. Ducal Prussia consisted of the historical regions of Lithuania Minor, Sambia, Upper and Lower Prussia, as well as Masuria8. Together with Warmia, these territories made up the province of East Prussia. Prior to 1795 though, the region of East Prussia had already been steadily increasing its German settlement, due to the previously-mentioned forces of the Ostsiedlung.

This settlement, first driven by the Teutonic Order, led to the German displacement of the native people known as the Old Prussians. By the early 18th century, war, conquest, assimilation, and disease had eliminated these people, surviving only in the toponyms that remained. The German settlers mixed with Baltic Prussian substrate and bumped against the Lithuanians further east.

Much of northeastern East Prussia was, up until 1945, known as Lithuania Minor, due to the strong historical Lithuanian heritage in the area. By 1900 though, much of the native Lithuanian population, like the Baltic Prussians before them, had been assimilated thoroughly into German culture. This started much further back though, when in 1724, by royal decree, all Poles, Jews, and Samogitians (the contemporary term for Lithuanians) were forbidden from settling in Lithuania Minor to make way for German colonization9.

While most Lithuanians today are solidly Catholic, the Prussian Lithuanians, or Lietuvininkai, strongly identified with Protestant German religion. A similar story exists for Masuria, where Masovian Polish settlers came to identify with the German nation they lived in. Thus, the land of East Prussia lay at the intersection of Baltic Prussian, German, Polish, and Lithuanian cultures.

III. Linguistic and Cultural Minorities

Looking at the figures for minorities over time, there are a few that need to be addressed. Given the role of the German census in inflating the number of Germans, as would be expected in the age of the Kulturkampf, we create estimates for what the actual population would have resembled.

A. Lithuanians

The first such minority language to be addressed is Lithuanian, spoken primarily in the northeasternmost parts of East Prussia10. The overall region of Lithuania Minor had been strongly Germanized by 1905 - according to the census only a strip of land north of the Neman river still was majority-Lithuanian. While it is undoubtedly true that Germanization had stripped much of the native Lithuanian population of their erstwhile language, some of these numbers seem inflated.

Since East Prussia had been banned from settlement by non-Germans in 1724, changing demographics can only be attributed to Germanization and German colonization. Many people from East Prussia moved west in the later 19th century to look for work. These language changes over time therefore most likely represent autochthonous Prussian Lithuanians, or Lietuvininkai, being counted as Germans. Exactly how many remains a question - unlike the Polish language elsewhere in eastern Germany, the Lithuanian language was not forcibly suppressed.

Looking first at the numbers from pre-German Empire censuses, we understand the clear picture of Germanization. Given that nearly 40 years passed between the censuses, and life expectancy was still increasing, it is not unreasonable to infer a large degree of population churn, in which older bilingual German-Lithuanian speakers were replaced by younger monolingual German speakers. Particularly as Germanization made German the only language used at school and in official public business, it is reasonable to see how a language could change this quickly.

However, before the advent of 20th century’s fascist ideologies and modern technology to propagate them, languages did not simply disappear that quickly without a critical reduction in population. As long as a language has a critical mass of speakers and maintains that critical mass of speakers, it will survive. The Polish language survived over 120 years of state-sponsored repression. Lithuanian did not undergo nearly the same, particularly under the relatively lenient Prussian dominion. Thus, remedying a slight undercount by revising higher the census counts of Lithuanian speakers is likely to hew closer to the true estimate of the actual population11.

German ethnographer Franz Tetzner surveyed Lithuania Minor in 1898, and noted that the language ranged in five different bands12. Furthest north was the Memelland, or the area of Lithuania Minor immediately outside the city of Memel. This region had the most Lithuanian speakers. Further south, a second band between the Neman river and the first. Below that lay the swamps of the Elchniederung, which had already undergone significant Germanization. Two more bands lay below, as far south as Insterburg and Stallupönen, but by the 1890s these areas were widely understood to be heavily Germanized13. When comparing the data to that compiled by Lithuanian linguist Friedrich Kurschat in 1876, we can clearly see that the boundary line for the Lithuanian language has moved quite far to the north. Thirty years before our 1905 census, sizable Lithuanian-speaking populations existed as far south as the Pregel and Inster rivers14.

By comparing Tetzner’s notes with the official data from the census, we can infer where the census counts were likely off and make our own estimates. Looking at the actual numbers as estimated, we find that the southernmost 5th region, stretching as far south as the modern Polish-Russian border, is in fact nearly fully Germanized as Tetzner indicated. Above that lay the 4th, which is estimated at 86% German and 14% Lithuanian. Again, this seems correct according to the data. North of that lies the villages of the Neman river, particularly in the Elchniederung swamps. Here we do not see much of an increase in the Lithuanian percentage, which is estimated at 16%. We would expect the gradient to be a little higher given the settlement pattern of new Germans coming from the south, and the proximity to Lithuania Minor. This is the first band that will see its Lithuanian numbers adjusted upwards slightly. As for the last two, the next band was 54% Lithuanian, and the next one after that was 66% Lithuanian according to the census. It seems that the northernmost 3 bands should have their counts revised upwards.

A table of the 5 bands as drawn by the census figures:

How then, should each band be shifted? According to Tetzner we know that in the first band, Lithuanian language was ubiquitous even as late as the 1890s - an additional 15 years would not be enough to wipe this out. Religious statistics are not a good clue Prussian Lithuanians were predominantly Protestant like their German counterparts. The solution is to look at what the pre-Germanization figures from 1867 at the Kreis-level, and come to a medium value. Obviously there would be some organic shifting between Lithuanian and German culture, but given that the population makeup is overall the same, the incentives that made thousands of Lithuanians identify as Germans on the census must be extricated from this analysis.

Looking at the rural Memelland, Heydekrug, and Tilsit, we can observe the more rural Heydekrug district and compare the statistics. This will be the best estimation of an error-free population that we can get. Using the 1867 population statistics which show the Memel district as majority-Lithuanian, as it would have been in that time, we see that a proportionate estimate of the Kreis-level population produces 16,358 residents in Memel city. Of these, we can estimate using the 1905 population schedule, discounted slightly, that 80% of these are Germans. Deducting this, we find the rural Memel district in 1867, before Germanization, to be an estimated 28% German and 71% Lithuanian. The numbers to be re-Lithuanianized most likely are found in the population-rich suburbs of Memel.

In the second band, we apply the same methodology removing the large more cosmopolitan city of Tilsit. We see that Tilsit has 40% of the overall population, overwhelmingly German. In 1867, that would have made the rural population of this band roughly 76% Lithuanian as well, fairly consistent with our initial estimates. Finally, in the third band that encompasses the districts of Pillkallen, Ragnit, and Niederung, with no major cities, we can compare the 1867 numbers to the 1905 numbers and estimate that the German percentage figures are roughly 10% larger than they ought to be. Thus, we have our targets for undoing the forced Germanization.

With the targets created properly, we created weights for each town and village based on population - with more highly-populated ones receiving higher weights. Naturally, more cosmopolitan cities which were predominantly German were excluded, as those areas would have more accurate census counts of who counted as German. The Lithuanian percentage in band 1 increased by roughly 4 percent, after conducting a weighted average of the 1867 pre-Germanization numbers and the 1905 actual census numbers.

Most of the northern Memelland is now dark-green and has absolute Lithuanian majorities almost everywhere, which is consistent with Tetzner’s accounts. In band 2, the shift was larger, as a district as rural as Heydekrug and Tilsit north of the Neman river was made up of more autochthonous subjects of the German Empire. The Lithuanian population was estimated 14% higher than it actually was reported in the census - and this shows the villages north of the Neman darkening significantly in their shade of green. Finally, the third band had the Lithuanian percentage increased by roughly 6% - and when put in conjunction with the rest of the map, the gradient looks realistic and presents an accurate estimation of the true linguistic situation.

The Lithuanian majorities dramatically and quickly fall south of the Neman river, notably in the western portion near the mouth of said river, in a swampy moorland known in German as the Elchniederung. This area was once sparsely inhabited and only received its larger population through recent German settlement, and so it makes sense that this area is nearly completely German-speaking, as Lithuanian was primarily the language of autochthons who did not move around.

B. Masurians

After the Lithuanians, the next minority group to be addressed are the Poles/Masurians. The largest Slavic-speaking populations in East Prussia are a curious case of language and culture not entirely lining up. The name “Masurian” comes from the Polish word describing a settlement of colonists from the neighboring region of Masovia. Masurians historically spoke a dialect of Polish, but were otherwise culturally Germanized.

As Richard Blanke points out in his landmark paper on the Masurians15, the closest equivalent to the Masurians known throughout Western Europe would be the Alsatian population in eastern France. Speaking a dialect of German, the Alsatians very clearly had no desire to become part of Germany, and were forcibly integrated, many times against their will, into the greater German cultural and linguistic community.

The extant literature on this subject makes clear the confusion felt by Polish activists in the 19th and 20th centuries regarding the apathy towards the Polish nationalist project. In our visualization, we take these perspectives into account and are electing to maintain the separation between the categories of “Polish” and “Masurian” that the census has maintained. However, even in a place as German-aspiring as Masuria, the effects of forced Germanization can be seen in the census numbers.

One reason for this is the historical. The territories known as Masuria were never part of Poland proper prior to 1795, when the Polish-Lithuanian Commonwealth existed. In contrast to Warmia, which *was* part of Poland proper and was once referred to as Royal Prussia, Masuria was part of Ducal Prussia, subject to the Teutonic Order’s rule and was thus not culturally imprinted with the same Polish heritage. Particularly as the cultural heritage of southern East Prussia stems from the Teutonic Order’s conquest and domination of the region in the 15th and 16th centuries, it makes sense that the culture and state set up by the Order would ascribe to German-dominated Protestantism.

As Franz Tetzner also writes, by the 19th century, many areas that used to be dominated by the Polish language (the Masurian dialect in all localities outside of the Soldau area and Warmia) had become Germanized. Polish was increasingly only used in church services aimed at older people16. This process, as Tetzner describes, occurred even before the forced Germanization of the 1870s, in that Masurians voluntarily began to identify as Germans. With the introduction of the railroads in the later half of the 19th century, the language declined some more, as a product of isolated Masurians making contact with Germans from other parts of the Empire through emigration and new visitors. Richard Blanke also mentions that German language was required in schooling - a decision that was not widely protested by the citizenry of Masuria, unlike in other parts of Poland.

According to Blanke, it was widely accepted by the German authorities and by Polish activists that the only thing Polish about most Masurians was their language. Looking at a political level, we can see that the Masurian districts had settled into being bastions of the German Conservative Party in Reichstag elections. As Blanke shows, candidates of the Polish Party never won any mandates in the Masurian districts, which contrasts with their success in other Polish-speaking regions, like Posen, West Prussia, and Upper Silesia.

This is further reflected by the 1920 plebiscite for Masuria. Held in a plebiscite zone that had a strong majority of Polish speakers, only 2% of the vote went to Poland - 97% voted to remain in Germany. Some of this could undoubtedly be to voter intimidation, as the pro-Polish side claimed at the time. But investigations from the Allied authorities leading the plebiscite found scant evidence, instead attributing the lopsided result to a lack of currency among the populace for the pro-Polish cause. A revote under fair conditions would have likely meant a similarly overwhelming result to remain in Germany. This is in stark contrast to the 1921 plebiscite in Upper Silesia, where a majority of Polish speakers *did* identify with and desire union with Poland, and where the results were much closer, contested, and plagued by violence17.

For the Masurian population in 1920, Poland looked like a completely different country, at the time beset by Soviet invasion, with worse economic prospects and an alien religion. After World War I, the Germanization process accelerated, with the Masurian language declining precipitously. During the era of the Weimar Republic, the Masurian districts gave the political right wing their largest vote share anywhere in the country, particularly in elections featuring the Nazi Party. And after 1945, most Masurians opted to move to Germany, rather than stay in Poland with their ostensible countrymen. Thus, even though by blood, Masurians may be Polish, by the daily choice of nationhood, Masurians clearly preferred to be identified as Slavic-speaking Germans.

According to Tetzner, the region would have been bilingual at the turn of the century:

“In both parts, sermons are regularly held in both languages, but in the northern part, due to Germanization, Polish services are increasingly rare. The language differs considerably from Polish, especially the dialect east of Kurken, Lindenwalde, and Seelesen on the eastern border of Osterode. Almost all residents speak both languages; there, Masurian survives far longer than Lithuanian. At school, the children learn German fairly well; once they leave school, they gradually forget it more and more. This is not due to an aversion to German. The Protestant Masurian is German and Prussian through and through, and strictly separates himself from the Catholic Poles.” As described in the past, the Masurians set themselves apart from their Polish cousins to the south - seeing themselves as having better cultural and economic prospects under German rule.

Now that we have identified Masurians as clearly being a separate group from Poles, the next question is assessing the impact of forced Germanization. According to official census statistics, in 1861, fully 74% of the population of the Masurian districts spoke a variant of Polish. By 1910, that number was down to 52%. Using more revised and accurate statistics from Polish historian Leszek Belzyt, that number is going down from 83% to 69%18. In other words, in 1861 there were 9% of the population that were counted as German but actually were Masurian. By 1910, that number had climbed to 17%. Clearly, we can observe the effects of census undercounting by the German authorities, and this must be remedied. Tetzner clearly states that the pace of Germanization was accelerating towards the latter half of the 19th century, with districts such as Rastenburg and Goldap, roughly midway between north and south in East Prussia, becoming purely German by the 1890s. In future, he writes:

“Humanly foreseeable, therefore, in a short time, Rastenburg, Goldap, Angerburg, and Allenstein will become purely German, while Loetzen, Marggrabowa, Osterode, and Sensburg will succumb to the Germanic influence advancing from the north. Ortelsburg, Johannisburg, and Neidenburg could remain Masurian strongholds for some time due to the Polish hinterland, the few railways, and the large forests, even though the district towns will soon be predominantly German. The city of Lyck, home to a high school and a garrison, can already be called a purely German city; The border parish of Ostrokollen has acquired a German character in a short time, especially since the expansion of the railway lines, so that the German population coming from the northwest already predominates along the Loetzen-Prostken railway line, i.e., up to the Russian border.”

Up until World War II, this prediction from Tetzner held up remarkably well - the districts which were the most rural and isolated kept their Masurian-ness the longest - namely, Ortelsburg and Johannisburg on the border with Russia/Poland. For the purposes of our analysis, this means that we will have to evaluate, on a Kreis-level basis, what percent of Germans are actually undercounted Masurians. The same goes true for Kreis Allenstein, which, while not formally part of Masuria, is attached as part of Regierungsbezirk Allenstein. Allenstein is part of Warmia, and so had proper Polish character due to its strongly Catholic nature. But like much of Warmia, the area had been thoroughly Germanized in the intervening century plus since the partitions. The Polish language spoken in Warmia was apparently different from the Masurian dialect, and thus should be counted separately. Polish dialects have always been rather mutually intelligible, and thus, it makes sense to combine all the non-Masurian dialects in the region as one Polish language.

Thus, having addressed the question of Masurians as an identity, and the degree to which this identity’s decline was the product of force, we must address the next step of our analysis - answering the question of how one defines Polish versus Masurian. We know language by itself is not sufficient to warrant the classification of one or the other. We see that in most of East Prussia outside of Regierungsbezirk Allenstein, the Polish population is significantly larger than the Masurian population. Looking at language and religion then makes more sense. Within the Regierungsbezirk though, Masurians are the overwhelming majority, with most Kreise having upwards of 85 percent of their Polish speaking population listed as Masurian. In our future analysis, we will maintain these relative splits as they are in keeping with what Tetzner and Blanke have written - the Catholic presence tends to be non-native to the region, particularly in the Soldau area near Neidenburg in southwestern Masuria.

Tetzner in his analysis classifies any Pole or Masurian with Protestant religion as a “Masurian”. This is a solid basis to start, however, there were not insignificant numbers of Catholics who identified themselves as “Masurian”. In our analysis, they will be included in the final tally of the numbers, as they self-identified this way. Given that the prevailing cultural dynamic in the eastern portions of the Reich were between Protestant Germans and their Masurian allies versus Catholic Poles, a Catholic specifically declaring themself as Masurian should be noted.

From a linguistic perspective, the dialects of Polish spoken in Kreis Allenstein are different, but not completely alien. Linguistics classify both Warmian and Masurian dialects of Polish language as part of the Masovian category. However, due to the stronger Catholic roots in Warmia, Germanization was in fact more forced and the Polish language will have a stronger incentive to be underreported in official census data. Thus, in splitting Poles and Masurians, we will take count Catholic identified as Polish-speaking, as Polish. Any Protestant Pole, or any Masurian of any religion, will be counted as Masurian - even the Catholic ones who self-identified as Masurian (these are no more than 5% of the Masurian population in any Kreis with significant numbers).

With correcting the statistics, unlike in the case of the Lithuanians, it takes more effort as we have been provided Kreis-level estimations for the actual number of Polish speakers in each of the major districts. Tetzner’s bands of Lithuanian speakers were cross-referenced against prior statistical numbers, but we have hard percentage estimates that have been vetted in the form of Leszek Belzyt’s calculations. In recalculating the estimates, we estimate that the Germanization process in districts Goldap and Rastenburg were nearly complete - the 1905 census data shows that less than 2 percent of the population spoke Polish or Masurian. This leaves the following districts to be amended: Johannisburg, Lötzen, Lyck, Neidenburg, Ortelsburg, Osterode, Rössel, Sensburg, Oletzko, and Allenstein.

The first step in calculating the revised numbers was projecting a corrected estimation for each district using a composite of prior census years and the Belzyt figures. Using a weighting where 75% of the 1910 census figures and 25% of the 1890 figures provided a reasonable estimate, on average roughly 10-15% more Slavic-speaking than the official figures would indicate. This meant that a district like Lötzen, famous for its Masurian culture, would actually be closer to 60% Slavic-speaking rather than the 44% that the official census figures denoted. Generally, the official map of the data is accurate in denoting which areas were German versus Polish speaking, however the official data definitely undercounts the amount of bilingual German-Polish speakers and places them into the category of Germans, which obscures the full linguistic picture. For the districts where Belzyt’s data is not figured, namely, the Warmian districts of Allenstein and Rössel, a target figure is estimated by looking at the pre-Germanization figures and creating a reasonable estimation. In Allenstein, the 45% Polish-speaking population rises to 53%, and in Rössel, the number barely changes much, given that virtually 90% of the district’s area had already been Germanized.

The first step was to figure out the magnitude of shifts. Using a baseline that areas with more Germans had more room to shift, a preliminary shift factor was done at the village level, to create a new tentative German-Polish linguistic split. After that, the remaining number of people to be shifted were allocated by the percentage of the German population that lived in each village, excluding a few larger cities that would have been accurately described in the census as purely German. Once the binary German-Polish split is calculated in each village, then the Polish/Masurian numbers are calculated by deducting German population and adding it back to the Masurian population (Polish population in the Allenstein and Rössel districts). The result is something closer to what the qualitative sources describe - the cities are strongly German but the rural villages surrounding them predominate with Masurian language. The Polish population that is Catholic only exists strongly near Allenstein, and in a few scattered villages near the border with West Prussia, which had undergone much less Germanization and were much closer to Catholic Polish culture.

The statistics show that between Germans, Poles, and Masurians, the Polish language was still dominant in all of the districts of Regierungsbezirk Allenstein. Germanization was the strongest in Kreis Oletzko, in the far-eastern part of the region, with 47% of the population counted as German (down from 66% in the official census records). As described by Tetzner, the rural isolated heartland of far southern Masuria (districts Neidenburg, Ortelsburg, and Johannisburg), had upwards of 70% Masurian language speakers, which combined with the Polish Catholic population to maintain nearly 80% of the population as Slavic. In Kreis Allenstein, visible on the map are pockets of Masurians, in what is obviously a Warmian land. Protestant plurality villages in a strongly Catholic area could perhaps be German settlers, but it is much likelier that the linguistic proximity between the Warmian and Masurian dialects of Polish leaves it easy for “Germans” to be disguised Poles in the census figures.

Of course, the 1905 census is probably the last time that such a census can be taken to show the stark isolation of the Masurians - in the next four decades the assimilation and wars would bring the Masurian experience much closer to the German experience. The Germanization that at first was voluntary and then began to quicken soon rapidly accelerated further.

By 1945, when the German presence in the region finally ended, there was not an identifiable Masurian language-speaking community left among any but the very oldest of church parishioners. Language and culture are not set in stone, but 1905 data before the wars is the closest one can get with official data to finding a hypothetical “State of Nature” for the sociolinguistic situation in the region. But this analysis is a faithful attempt to recreate the extent to which Masurian culture had dominated southern East Prussia right until the turn of the century.

C. Latvians

After the Lithuanians and Poles, there is an additional case: the Baltic people living on the Curonian Spit known as the Kursenieki. These people lived on the shores of the Curonian Lagoon and the Baltic Sea between Palanga and Danzig originally. Their language, known in German as “Nehrungskurisch”, is described as a western dialect of the Latvian language. Due to centuries of assimilation and encroachment from neighboring German and Lithuanian peoples, by the turn of the 20th century, the Kursenieki people only numbered a few thousand at most.

The language they spoke, called Nehrungskurisch in German, was more commonly used among coastal villages as a trade language among the fishermen that lived near the Curonian Lagoon. Nehrungskurisch is widely seen as a dialect of Latvian, a Baltic language related to Lithuanian. But given the close proximity to Germany and Lithuania, many Kursenieki were bi- or even tri-lingual in their own native tongue as well as German and Lithuanian19.

The German censuses never collected official data on this language, likely due to this particular point. During the era of Germanization, non-Germans who had German-speaking ability were in many instances counted as German. It is reasonable to assume that Latvian-Lithuanian bilinguals were counted as monolingual Lithuanians. Until 1945, the Curonian Spit was a very isolated region and so the Kursenieki community remained small and insular as its own identifiable community.

Parallel to the censuses conducted every 5 years, there exists a study of Slavic peoples in the Eastern territories of the German Empire, the details of which were recorded in Tetzner’s book. Tetzner traveled to the Curonian Spit to conduct research on this population in 1897. By this time, much of the southern Curonian Spit had been Germanized - due to Kreis Fischhausen’s proximity to the provincial capital of Königsberg. In the settlements of Sarkau and Pillkoppen, less than 10 Latvian speakers existed. Further north in Nidden, Preil, Perwelk, and Schwarzort, the distance preserved the language in a better state. At the northern tip of the Spit lay the settlement of Süderspitze, which was founded as a suburb of the larger city of Memel. Süderspitze, as well as coastal settlements further north were not as heavily populated with Kursenieki.

According to Tetzner’s map, the 10 settlements with notable Kursenieki populations were, from north to south along the Spit: Nimmersatt, Karkelbeck, Mellneraggen, Bommelsvitte, Schwarzort (municipality and manor district), Nidden, Preil, Pillkoppen, and Sarkau. Tetzner collected data of both the Latvian population themselves, as well as the overall number of speakers of their language - namely, the aforementioned German and Lithuanian fishermen. For the purposes of this analysis, we will only count the Latvian number for our Kursenieki analysis. By Tetzner’s calculation, there were 1196 Latvians living near the Curonian Spit. Given that his survey was conducted in 1897, we must create realistic estimates for the population in our analysis for 1905.

We achieve this by finding the 1895 census populations for each of these settlements, and calculating the population growth/decline to the 1905 census figures. The 1897 figures from Tetzner are best benchmarked against the 1895 numbers - so the question becomes, how did the Latvian population increase relative to the overall 1895-1905 transition? We begin by examining each settlement and examining the conditions for each.

Starting in the south, Sarkau and Pillkoppen in Kreis Fischhausen both increased their populations dramatically between 1895 and 1905. However, as mentioned, given the proximity to more heavily German localities like Pillau and Königsberg, it is highly likely that this new population growth was powered by German in-migration. In 1897, there were only 6 Latvians in each settlement - mostly old fishermen who were unlikely to increase their numbers in 8 years. Conservatively, we estimate those 6 in each settlement stayed the same.

Further north, in Nidden, this was in fact a more heavily Kursenieki settlement with a younger population base, but around this time a German artists colony came into being. Given the high population growth, we decided to attribute half of the overall growth to German in-migration and half to native Kursenieki growth. That means that the 564 Latvians that existed in 1897 now became 605. Preil and Schwarzort observe similar dynamics but do not have the same German influx, due to the lack of an artists’ colony. Thus, we attribute 60% in Preil, and 65% in Schwarzort to the growth of Kursenieki.

In Schwarzort, given the division of the area into a municipality and manor district, we split the 1895-1905 population growth proportionally between the two. Finally, in the northern villages of Mellneraggen, Bommelsvitte, Karkelbeck, and Nimmersatt, we apply the straight-line population decline to the Kursenieki population as well. These areas were not as isolated, and as such, it is entirely reasonable to see these numbers decline with 8 more years of Germanization. In summation, whereas Tetzner calculated 1196 Latvians as living in the area in 1897, our analysis estimates roughly 1300.

With completed estimates for the 1905 Latvian population, the next task is to figure out which ethnic group to deduct the population from to create the new demographic in the data. Again, a municipality-by-municipality approach is necessary here. In the far north in Nimmersatt, given that the area was overwhelmingly Lithuanian and served as a German border town, the Latvian population was deducted from the Lithuanian. This was the same for Karkelbeck, Mellneraggen, and Bommelsvitte as well.

For Schwarzort, it made more sense to deduct from the German population given the isolation of the area, and that between the two languages, Lithuanian was definitely more widely used in the northern Spit villages - thus the German as recorded in the census was artificially high. For Preil, the census recorded nearly all residents as Lithuanian, which made this the obvious choice to deduct. The Nidden artists colony, as well as the growth in German population as recorded by Tetzner in recent years, meant that the German totals were likely artificially too high. Finally, for Sarkau and Pillkoppen, virtually zero residents were recorded as Lithuanian, so the German population was the obvious choice for deduction.

When all is calculated, we find that the Latvian-Kursenieki population is a plurality or majority all throughout the northern Curonian Spit. The villages of Nidden, Preil, and Schwarzort all have the Kursenieki as the largest population. This is corroborated by a 1907 map from Langhans, which shows a similar dynamic20.

One possible source of data error here is the presence of bilingual residents recorded in the German census. But given the relative lack of clarity provided by the German census here, it is better off and more feasible to construct the Latvian demographic from the monolingual columns of the dataset. There is no guarantee of which languages are included in the bilingual column - thus, leaving it as is remains the best option to produce a faithful result that is consistent with qualitative descriptions of the Curonian Spit’s linguistic situation.

D. Russians

The last major minority to talk about is the group of “Philippons”, descended from Russian Orthodox Old Believers. This population descends from a group of religious dissidents invited to settle in the Masurian forest by the King of Prussia in the early 19th century. The isolated location helped preserve their culture well, and they built several edifices and monuments to their Russian Orthodox faith. By the turn of the century though, according to Tetzner, all non-German citizens were expelled from the area, leaving a smaller population of Philippons. According to Tetzner, by 1899 there were roughly 500 such Old Believers in the vicinity of the town of Eckertsdorf in Kreis Sensburg. Roughly ten original villages were settled, which by the time of the 1905 census had been condensed into eight settlements.

In estimating the number of Philippons, we first look at the 1910 census, which has information on “other language”. In Eckertsdorf, 287 people, nearly half the population, are listed as speaking a language other than German, Polish, or Masurian. This is likely the Russian-speaking Old Believer population. Similar numbers can be obtained for the remaining villages. Now, looking at our 1905 census data, we can see that the 1910 “other” category closely approximates the summation of all Catholic believers here. When looking at Tetzner’s own categorization as well as the notes from the beginning of the 1905 census Gemeindelexikon, we can see that the category of “Catholic” included Orthodox Christianity as well.

Thus, for the purposes of this analysis, we combine all Catholic believers into the new category of “Philipponen”. Given that this is deep in the Masurian jungle where Germanization was deeply effective, there is not a large amount of evidence to support a sizable Catholic presence in the percentages described in the census. Perhaps in the nearby town of Ukta which was less isolated and secluded in the forest yes, but in the case of the Philippons, we see that their isolation kept their culture alive. The map shows that beyond Eckertsdorf, the Philippons did not greatly predominate. On the Standesamt map, the Russian presence is barely visible - most villages near Eckertsdorf had roughly 20% of their population as Philippons.

Thus, with these main minorities addressed, the process required to actually generate the maps can be described further in detail.

IV. Map Creation Process

With respect to the creation of the actual map, a number of sources were used. The most important method was combining the raster maps from the Meyers Gazetteer, along with latitude and longitude information gathered from the website genealogy.net. Each district had its gemeinde (communes) drawn from the map, with the boundaries drawn imperfectly, as there are no extant maps of the full boundary set.

For the part of East Prussia that lies within modern Lithuania, the place names are mostly Lithuanianized variants of the preexisting German names, so identifying place names was not that difficult, except in cases where the names changed. Particularly towards the northern Memelland, many old Lithuanian villages had names of people, such as Wessat-Hermann and Gibbischen-Peter. Today, those villages have Lithuanian names that do not bear a resemblance, and so additional research was done on a case by case basis.

The total list of villages and cities was gathered first from the Gemeindeverzeichnis, an online source listing every village in the German Empire and its population. The actual list of villages underwent only minor changes between the 1900 date from the Gemeindeverzeichnis and the 1905 that we use for the census data. At first the method was rather inaccurate but once we figured out how to access downloadable Meyers Gazetteer maps as rasters in QGIS, each district was drawn and then cut into its respective villages.

Given that the latitude and longitude data was not completely accurate, a fair bit of guesswork and searching had to be done for the last few villages per district, particularly some of the “Gutsbezirke”, or estate districts. In most cases, Gutsbezirke shared names with Gemeinde that were nearby, and so they were drawn very close by for the sake of realism.

This map would not have been possible without the Meyers maps because once the digitization process crossed into modern Russian territory, the amount of cognate place names dropped significantly. Russian has only been spoken in Prussian lands since 1945, and it has wholly erased the Prusso-German heritage. For example, on the Curonian Spit, the town of Cranz/Kranz was called that in German. In Polish, it was Koronowo. In Lithuanian and Baltic Prussian, it was Krantas. The modern Russian name of the town is Zelenogradsk, with no particular similarity.

In Poland, the names mostly remain Polonized variants of the original German names, but particularly in the Masurian districts drawing the numerous lakes was an onerous task. Many such Gutsbezirke had lakes with names in German, that are today called by other names in Polish. The largest lake in Masuria is today called Śniardwy, but in German it was called the Spirding-See. For smaller lakes this was a much more laborious task.

East Prussia’s vast forests also presented a challenge - particularly as many were organized in districts labeled “Oberforsterei”. Serving as forest ranger stations, the decision was made to create very large township-like structures that covered the area of the forest as described on the map. The relative area of the polygons created might not hew exactly to the area numbers described in the census, but this is an impossible task without actual boundary maps. Our goal is to be as exacting as is practicable, and to infer based on the rest of the data that is available.

The overall aesthetic goal of the GIS project was to create a visualization of East Prussia similar to a New England town map. In a New England town map, towns of varying population densities generally have similar-sized areas. For example, the city of Cambridge in Massachusetts has roughly the same area as a town like New Hubbardston, even though Cambridge has roughly 60 times the people of the latter. This mostly comports with the general Prussian and pre-1945 European model of towns, where settled areas are roughly evenly spaced out between large fields and forests.

When viewed in aggregate from a high level, it largely accurately models East Prussia in the same settlement style as the rest of Germany. The only area that was subject to major suburbanization and annexation was the city of Königsberg, which had started to annex several villages in the immediate vicinity starting in the late 19th century. This was, naturally, the largest city in the whole province. Other cities, such as Gumbinnen and Insterburg, did not have any identified villages around them for a few miles, but in the interest of aesthetics, were kept to a smaller size to comport with the look of the neighboring villages around them.

When all was done, a final correction was made by going through the Standesamt, or civil registration office, list, to make sure that all the polygons had relative geographic contiguity. It was in this step that the Lithuanian villages drawn by hand without the map were corrected to be in their proper places. Roughly 7400 villages, towns, and estate districts were reduced to about 750 Standesamt. These could roughly be approximated by the Polish gmina, which is the smallest administrative unit of government composed of a few villages. The word gmina, means commune, and comes from the same root as the German Gemeinde - even though it has a different functional meaning. The Standesamt boundaries have never been digitized for East Prussia, and this likely represents the first time that all the available resources have been combed to resurrect the geographic scope of the former province. 80 years later, the full picture of the cultural narrative of East Prussia comes into view.

V. Results

The first and most apparent observation is the extent to which Germanization had affected the northern parts of East Prussia. Though nearly unanimously-German outside of the Memelland, the Baltic toponyms still survived well into the 20th century, until many were changed as part of the Nazi regime’s further Germanization policies. We can see that the toponymic names that bear Lithuanian features are found well outside of Tetzner’s extant Lithuanian region that he described at the turn of the century.

At a density-based view, we see that East Prussia was overwhelmingly rural. Outside of the small cities that dot the landscape, the vast majority of the towns had densities of less than 200 people per square mile. The contemporary comparison that works best would be parts of the peripheral American South. Unlike American rural areas attached to a metropolitan area, there is little economic productivity that incentivizes population growth and economies of scale. It is not, however, so sparsely populated in the way that much of the American West is. The presence of such a geographically disparate population ensured that the culture of East Prussia would be dominated by rurality.

Notable is the lack of any major urban agglomeration. Western Germany, all along the Rhine, is full of industrial cities and wide swaths of population. Even Königsberg, the largest city by far, is notable for a lack of suburbanization similar to what one would expect. Granted, at the turn of the 20th century, the economic forces that led to suburbanization were only beginning to take effect in more industrialized and developed parts of the world - but people who established new residences outside of the preexisting city limits mostly remained in small towns.

Königsberg may not lie on the Rhine, but the Pregel River was no small stream - it cut through all of the region, and connected the coast with the hinterland, and further on to Lithuania and Poland. In the pre-modern world, of which East Prussia’s geographic layout was certainly derived, economic productivity usually either came from urbanization-fueled industrial might, or from maritime trade. The lack of industry is already mentioned. Königsberg was certainly a city of note, but it was still much smaller than Danzig to the west, where the outflow of the Vistula led to the Baltic. The only other port city of note, Memel, was a much tinier city, not anywhere near the scale of a Cologne or a Düsseldorf in the German economy. Königsberg’s port activities were thus significantly less economically productive in absolute terms.

The economy of the region was therefore primarily driven by landholders exacting agricultural wealth - preserving an archaic quasi-feudal economic system reminiscent of the American South. It is no coincidence that, just like East Prussia, until very recently with the diversification of the economy, the South was long the poorest part of the United States.

When one looks at income at a village level, the picture becomes even more interesting. The census provided property tax data for each settlement, per unit of area. By multiplying by the area and dividing by the population, we get a useful proxy for GDP per capita at this granular level. We see that East Prussia overall is very poor, with most villages having a GDP per capita of between 20 and 60 Reichsmarks.

The most immediate thing to notice is that the cities and small towns distinctly are poorer per capita than the surrounding settlements. Looking at the settlements with densities of 500 people per square mile or higher, virtually all of them are in the poorest bracket of incomes. This makes sense given that income inequality was comparatively smaller in urban areas. This would not have happened in cities, where people’s wealth would have been on average comparatively lower.

Most rural areas in East Prussia were dominated by the rich landowning Junkers, who let poor liberated serfs live on their land to work. Given the law of averages, the extremely high land value that was enclosed in all of the estates skewed the rural areas massively towards being higher. Furthermore, this is confirmed when you look at Gutsbezirke, the Estate Districts. The color difference is stark - the Gemeinde municipalities are shades darker than the Gutsbezirke.

Prior to their abolition in the 1920s under the Weimar Republic, estate districts had their own independent taxation schemes that involved extra money being exchanged not between a citizen and the Empire, but accounted for obligations between citizen, Empire, and noble landlord as well. These maps therefore show the difference with stark precision:

Looking at these maps a few regions stand out. East Prussia in general was poor, but areas with higher concentrations of ethnic minorities were even poorer. In the north, Lithuania Minor was very poor, particularly towards the border with Lithuania proper in the Russian Empire. It is less apparent, however, as you see a line of less-indigent villages heading northward towards Nimmersatt, that are likely the product of in-migration along old familiar routes. Away from the road, development by the German authorities was probably not as likely, and thus the areas remained poor. The Masurian districts were. to a village. all dark blue.

This can be attributed to the fact that Masuria was even more isolated and undeveloped than Lithuania Minor, which had commerce crisscrossing the land. Masuria did have trade and transport from Poland, but given how forested the region was, and how un-Germanized the region was until quite recently, it is no surprise that development came very late to the region.

Similarly, on the Curonian Spit, the Kursenieki-speaking areas remained poor fishing villages, while the more Germanized towns such as Bommelsvitte and Sarkau were comparatively wealthier. Though today coastal living is often associated with wealth, that was definitely not the case in East Prussia in 1905. The relative isolation of the Curonian Spit as well as the lack of a mobile economy led to the stagnation of the coast - most of these villages remained poor right up until the major expulsions in 1905. Further inland on the Sambian peninsula were a few Gutsbezirke that were comparatively wealthier - perhaps close enough to Königsberg that they served as manorhouses, similar to the function the Hamptons serve to New York’s famous today.

One notable exception is the swampland drained at the mouth of the Neman River, termed the Elchniederung. This can be explained by the fact that the population here was not made up of native Germanized-Lithuanians, but rather came to the region after German hydraulic engineering drained a lot of the swamps and made the area more habitable. Even on the linguistic map, the area is notable for being monolithically German in a region that hewed closer to an 80-20 split rather than a 95-5 split. In the income map, the Elchniederung villages are some of the wealthiest Gemeinde in the entire province. As a sidenote, it is a testament to the destruction wrought by the war and the aftermath that virtually all of the infrastructure necessary to sustain the population in these swamps was destroyed, and the area has returned to its swamplike roots.

The final note on the income map is the presence of Oberförsterei districts, or Forestry Office Districts. These are predominantly rural and wooded, and substantially wealthier than anything around them. This makes sense, as these are sparsely populated, and the owners of these districts would no doubt possess fabulous wealth in order to maintain their property. These are visible as the large green, yellow, and orange districts on the map that dwarf the surrounding villages.

VI. Echoes Today

Eighty years after the fall of East Prussia implicitly poses the question - how would the province fare had it survived to the present day, without the mass expulsions that actually happened? We look at preexisting electoral data and make realistic inferences about voting behavior patterns to see how a hypothetically reconstituted East Prussia would vote in German elections today.

During the German Empire’s 47-year period, East Prussia solidified into a bulwark of conservatism. At the outset of the Empire in 1871, the German Conservative Party (DKP) easily won most of the seats, except for the Catholic Center Party in Warmia and a few liberal victories occasionally. But by the mid-1880s, the dynamic was almost uniformly Center-leaning Warmia versus Conservative dominance elsewhere. This is not to say that there were no other parties. In the 1890s, members of the Polish and Lithuanian minority parties managed to secure wins in their home bases.

Furthermore, the city of Königsberg long remained a rare bastion of center-left politics. The liberal parties as well as the leftist Social Democratic Party enjoyed quite a bit of success. In fact, in the 1890s, Otto Braun, who would go on to become an SPD luminary in the Weimar era, ran unsuccessfully in more rural seats after having grown up around Königsberg. In summary, East Prussia’s mix of conservative, centrist, and liberal political history can be varied and diverse, but routinely was averse to leftism, outside of Königsberg. In the Weimar era, apart from the crisis years of 1919 and 1920, right-wing parties dominated always outside of Königsberg, except for Kreis Gerdauen. By 1932, when Nazis had consolidated the entire right-wing vote from the DNVP, Gerdauen alone remained pro-SPD.

At a geographic level, we maintain the 1914 borders of East Prussia for convenience sake, and to make maximal use of the available data. Thus, we are left with the issue of having many non-Germans in the simulation, an issue that has no real precedent in modern Germany that has arisen from mass expulsions and population transfers. We conclude that based upon imperial and Weimar results, Masurians would be staunch supporters of the far-right, and that Poles would remain solid with Catholic-leaning. Lithuanians would lean right wing but if poor enough, would have their own autonomist party, similar to the Danes in Schleswig-Holstein.

Between the Germans, with the Catholic vs. Protestant divide showing up greatly, we would see a center vs. right dynamic. When combined, the demographics of East Prussia suggest that it would follow an electoral trajectory of much of rural East Germany, that is to say, heavily leaning to the far-right. Had East Prussia stayed with Germany, it almost certainly would have been administered as part of the German Democratic Republic, and its population would have followed the same road to far-right populism that areas like Thuringia and Pomerania have seen. East Prussia was, and would be, even poorer than the actual East Germany, and even more susceptible to the far-right.

For the purpose of our model, we have the mainstream parties: CDU, SPD, FDP, Greens, and the Linke. On top of that, we add in a few regional and autonomist parties - the ECSU (Ermlaendische CSU), which would hypothetically mirror the role that the CSU plays in Bavaria, as a Catholic subsection of the overall Christian-democratic project. Each of the Lithuanian, Masurian, and Polish minorities have their own small activist parties that are exempt from the 5% threshold modeled. Finally, we add in the Alternative for Germany as well as the Heimat/NPD party to account for the far-right that would likely ascend.

The full model needs some more tweaking but in all realistic scenarios, out of the 87 seats in the pre-1945 Provincial Assembly (Provinziallandtag), the AfD and NPD combine for at least 30 - making this province likely a political basket case. There are three main scenarios, all involving the strength of the AfD and NPD. Example parliamentary groupings are created below, evaluated against the strength of the CDU and SPD.

Scenario 1: AfD and NPD do not form a plurality against every other party

In this case, the AfD and NPD are likely below 40% of the seats, which means that a workable CDU-SPD grand coalition would form in the state. There are several states today (Berlin, Hesse, and Saxony) that have CDU-SPD coalitions. While transformative legislation might not be passable with such an arrangement, the overwhelming right-leaning nature of East Prussia means that the CDU will likely be the senior partner in the coalition.

Scenario 2: AfD and NPD are the largest bloc in the Provinziallandtag, but fall short of a majority

In this case, the anti-democratic far-right bloc is only a few seats shy of a majority. Every additional party opposing the AfD and NPD would have to compromise on a lot to create a working governing arrangement. In this particular example, the socially conservative BSW party would have to be included in government. Due to the nature of government in Germany, it would be fairly difficult to bring down, but nonetheless would be highly unstable. A state government with an SPD-CDU-BSW coalition would be similar to the one in Thuringia, where the AfD recently became the largest party but is not in government due to the cordon sanitaire. Pretty much, the main purpose of the government would be keeping the lights on.

Scenario 3: AfD and NPD gain an outright majority of seats

This scenario is the one that would be the scariest. As East Prussia was poor and agrarian, it would follow in the steps of other similar areas of the former East Germany, had it survived to the present day, in being marginalized and left-behind, and a haven of far-right politics. The closest analog would be the area of Vorpommern, which then and now has been highly conservative. The AfD swept the elections there - and in this case, East Prussia would have an even stronger showing for the far right. The AfD and NPD, in the case that they take a majority, break 80 years of precedent and usher in a far-right government for the first time since World War II. No amount of protesting, cordon sanitaire, or coalition discussion could prevent the AfD and NPD from running roughshod over democratic norms in this scenario. Even a minority government here from the SPD and CDU would need support from at least the AfD on key issues, which would cause a crisis of legitimacy nationally in Germany.

VII. Addendum

This is the first full-scale dataset exploring the pre-1945 demographic and cultural history of East Prussia. The data that has been available is now made ready for analysis. This chapter in European history which is not talked about enough, 80 years after the end of World War II, now gets its due.

On a moral note, it is important to realize that competing narratives can exist at the same time. The population of East Prussia did give rise to extreme far-right ideology, as they overwhelmingly supported the Nazi Party. This moral evil in turn gave rise to another moral evil, the ethnic cleansing of centuries of culture and history. The small but robust Jewish community in East Prussia was also eliminated as a result of the hateful ideologies that swept Europe in the late 19th and early 20th centuries.

The Nazi Party undoubtedly had its strongest support in the eastern territories of Germany, among which was East Prussia - however, it is immoral to ascribe blame for the Nazis to every single resident of East Prussia, or to say that they deserved the atrocities committed. Up to a third of East Prussians opposed the Nazis and the far-right, and they bore the same consequences as much as the loudest Nazi cheerleaders. Wars of aggression for irredentist claims are evil acts, whether they were committed by the Germans against every single population of Europe, or by the Soviets against everyone they came into contact with.

Today, this lesson must be heeded, as Russia, greatest heir of the Soviet Union, seeks to repeat in Ukraine what they perpetrated in East Prussia - the mass destruction of a people, culture, and place, followed by the deportation of autochthonous residents and colonization with loyal Russians. This is a reversal of what the Nazis partially succeeded in accomplishing in Poland - and exactly what the Russians have done in Crimea more recently. By importing hundreds of thousands of loyal subjects with no indigenous ties to the land, they sever any link to the past and destroy any base for future irredentist claims against their imperialist projects.

This project began as an answer to the basic question of how German the character of East Prussia was prior to 1945. All of the history makes it clear that the loss of East Prussia then would be comparable to the Rhineland being emptied of its German population today - it was an integral part of Germany. The Polish section of former East Prussia has had its historical heritage preserved somewhat better - most of the major villages are still intact with Polonized variations of their German names - and connected to the EU, one can see that the new population has made its home in the old.

By contrast, the Russian Kaliningrad Oblast has torn down the old German heritage, with all the name changes and the thousands of villages not rebuilt after the war. There is no continuity between German East Prussia and Kaliningrad Oblast the way that one can argue there is in the Polish portion. The entire exclave serves as a military staging point for Russia to threaten peace-loving Europeans with its nuclear arsenal - cultural preservation is not a goal of their presence, military domination is.

This project is therefore an exercise in memory of a people and place who, because of their place and time, were wiped off the map. It is a testament against ethnic cleansing, and a testament against displacement. Though the survivors of the great expulsions themselves are dwindling in number, their memories will live on.

VIII. Acknowledgements

Special thanks is owed to so many people - including Uli Schubert from the Gemeindeverzeichnis project, without whom the listings of municipalities would not be able to be cross-checked. Thank you to the curators of the site Genealogy.net, which provided me with the geo-location of each and every village that was used to draw the polygons. Thank you to the Internet Archive, without whom the clarifying data from Franz Tetzner on each linguistic minority would have been significantly less accessible. Thank you to Angel Ortiz, Michael Foley, and Thomas Nowicki for talking through methodological decisions on how to best present the data.

IX. Data downloads

I have attached the full municipality-level data of the East Prussian Census of 1905 in Excel, both the original file as well as the amended estimates post-research. A GIS shapefile that matches both of them, as well as civil registration boundaries, and the administrative districts are also attached.

Data link: download from Google Drive here

X. Please subscribe and share!

Subscribe now

Share

Leave a comment

As an American, my viewpoint focuses on the United States. I have mapped hundreds of elections and aimed to understand the why of American political coalitions. By understanding the elections of the past, we can understand how and why the Democratic and Republican parties’ priorities came to be, and how that influenced their trajectories in the more recent past.

Similarly, I have begun to look into data from Europe and elsewhere outside the United States. Together, we will go into the data and tell stories that have not been told and bring to life old tableaus of history. Given my work with current election data, I feel that my old posting space on Twitter has outlived its purpose. Particularly, as the reality of the site’s ownership has changed the userbase, I feel that the historical analysis deserves a new home where longer-form content can flourish. This Substack will be a digital museum of cartographical history. The maps and visualizations themselves are interesting to look at from an aesthetic perspective, but also because they tell stories of people that inhabited places and took decisions.

Our first post will go live soon. As a hint - May 2025 is eighty years since the final fall of the old central European empires…

And in the meantime, please subscribe and share!

Subscribe now

Share Shadow of the Past

Leave a comment